The primary purpose of this grant is to train the candidate to be able to independently derive computational and theoretical models of laryngeal diseases, to verify these models by experiments in animals, and to use the results of theoretical and animal models to design new medical and surgical therapies for diseases of the larynx. Coursework in fluid mechanics, computational flow dynamics, vibrations, and acoustics will give the candidate the necessary knowledge to derive new theoretical models for laryngeal diseases. In addition the candidate will develop his own computational model of airflow in the canine larynx during phonation. The coursework will be supplemented by tutorials in speech science given by the candidate?s mentor, Dr. Ron Scherer. Experience in verifying theory in animal experiments will be obtained in the research part of this proposal. For his research, the candidate will use the flow visualization method he has already developed to determine the patterns of airflow in the larynx during phonation. The pattern of airflow in the larynx is important because flow produces forces that drive vocal fold (cord) vibrations, and certain airflow patterns may also produce sound by mechanisms other than vocal fold vibration. Current computational and theoretical models make assumptions about the flow patterns that have not yet been experimentally verified. The investigator will use flow visualization and the particle image velocimetry method to determine whether flow separation occurs in the glottis, whether vortices occur above the glottis, and whether the flow directly above the glottis is laminar. The candidate will then compare his experimental findings with current theoretical and computational models and with his computational model. Training in recognizing clinical applications of the work will be fostered by consultations with two prominent laryngologists, Dr. Gerald Berke and Dr. Randal Paniello.